[bug] The newest version of dpgen (0.10.6) can't compatible with newest version of deepmd-kit (2.1.5)

[gchenustc]

Bug summary

the default parameters in param.json will be set in a wrong way while training. For example, the option "set_prefix" will be wrongly added into the subsetction of "training" tag, but it's the subsection of "train"-"traing_data" tag. it raises error:
dargs.dargs.ArgumentKeyError: [at location training] undefined key set_prefix is not allowed in strict mode

DeePMD-kit Version

2.1.5

TensorFlow Version

2.9.0

How did you download the software?

conda

Input Files, Running Commands, Error Log, etc.

params.json for dpgen's input

{
    "type_map": [
        "N"
    ],
    "mass_map": [
                14
    ],
    "init_data_prefix": "../init/",
    "init_data_sys": [
        "cgN.POSCAR.01x01x01/02.md/sys-0008/deepmd"
    ],
    "sys_configs_prefix": "../init/",
    "sys_configs": [
         [
            "cgN.POSCAR.01x01x01/01.scale_pert/sys-0008/scale-1.000/00000[1-5]/POSCAR"
            ],
         [
            "cgN.POSCAR.01x01x01/01.scale_pert/sys-0008/scale-1.000/00000[6-9]/POSCAR",
            "cgN.POSCAR.01x01x01/01.scale_pert/sys-0008/scale-1.000/00001*/POSCAR"
            ]
    ],

    "_comment": " ***** train ***** ",
    "numb_models": 4,
    "default_training_param": {
        "model": {
            "type_map": [
                "N"
            ],
            "descriptor": {
                "type": "se_a",
                "sel": [
                                        8
                ],
                "rcut_smth": 0.5,
                "rcut": 5.0,
                "neuron": [
                    120,
                    120,
                    120
                ],
                "resnet_dt": true,
                "axis_neuron": 12,
                "seed": 1
            },
            "fitting_net": {
                "neuron": [
                    25,
                    50,
                    100
                ],
                "resnet_dt": false,
                "seed": 1
            }
        },
        "learning_rate": {
            "type": "exp",
            "start_lr": 0.001,
            "decay_steps": 100
        },
        "loss": {
            "start_pref_e": 0.02,
            "limit_pref_e": 2,
            "start_pref_f": 1000,
            "limit_pref_f": 1,
            "start_pref_v": 0.0,
            "limit_pref_v": 0.0
        },
        "training": {
            "training_data":{
                        "_systems":["system1_path", "system2_path", "..."],
                        "set_prefix": "set",
                        "batch_size": 4
             }, 
            "numb_steps": 2000,
            "disp_file": "lcurve.out",
            "disp_freq": 500,
            "_numb_test": 4,
            "save_freq": 1000,
            "_save_ckpt": "model.ckpt",
            "_disp_training": true,
            "_time_training": true,
            "_profiling": false,
            "_profiling_file": "timeline.json",
            "_comment": "that's all"
        }
    },
    "model_devi_dt": 0.002,
    "model_devi_skip": 0,
    "model_devi_f_trust_lo": 0.05,
    "model_devi_f_trust_hi": 0.15,
    "model_devi_e_trust_lo": 10000000000.0,
    "model_devi_e_trust_hi": 10000000000.0,
    "model_devi_clean_traj": true,
    "model_devi_jobs": [
        {
            "sys_idx": [
                0
            ],
            "temps": [
                100
            ],
            "press": [
                1.0
            ],
            "trj_freq": 10,
            "nsteps": 300,
            "ensemble": "nvt",
            "_idx": "00"
        },
        {
            "sys_idx": [
                1
            ],
            "temps": [
                100
            ],
            "press": [
                1.0
            ],
            "trj_freq": 10,
            "nsteps": 3000,
            "ensemble": "nvt",
            "_idx": "01"
        }
    ],
    "fp_style": "vasp",
    "shuffle_poscar": false,
    "fp_task_max": 40,
    "fp_task_min": 5,
    "fp_pp_path": "./",
    "fp_pp_files": [
        "POTCAR_N"
    ],
    "fp_incar": "./INCAR_scf"
}

input.json for training step (automatically generated by dpgen)

{
    "model": {
        "type_map": [
            "N"
        ],
        "descriptor": {
            "type": "se_a",
            "sel": [
                8
            ],
            "rcut_smth": 0.5,
            "rcut": 5.0,
            "neuron": [
                120,
                120,
                120
            ],
            "resnet_dt": true,
            "axis_neuron": 12,
            "seed": 4027317093
        },
        "fitting_net": {
            "neuron": [
                25,
                50,
                100
            ],
            "resnet_dt": false,
            "seed": 2450907653
        }
    },
    "learning_rate": {
        "type": "exp",
        "start_lr": 0.001,
        "decay_steps": 100
    },
    "loss": {
        "start_pref_e": 0.02,
        "limit_pref_e": 2,
        "start_pref_f": 1000,
        "limit_pref_f": 1,
        "start_pref_v": 0.0,
        "limit_pref_v": 0.0
    },
    "training": {
        "set_prefix": "set",    ########## error here! ############
        "numb_steps": 2000,
        "batch_size": 4,
        "disp_file": "lcurve.out",
        "disp_freq": 500,
        "numb_test": 4,
        "_save_freq": 1000,
        "save_ckpt": "model.ckpt",
        "disp_training": true,
        "time_training": true,
        "profiling": false,
        "profiling_file": "timeline.json",
        "_comment": "that's all",
        "training_data": {
            "systems": [
                "../data.init/cgN.POSCAR.01x01x01/02.md/sys-0008/deepmd"
            ],
            "batch_size": [
                4
            ]
        },
        "seed": 3242942795
    }

Steps to Reproduce

nothing

Further Information, Files, and Links

nothing

[HuangJiameng]

Strict check on run_param.json is loosen, see #952. Still, thanks for your report. It will be fixed.

[gchenustc]

Sorry, Because it's interrupted before, So it restarts and not read params.json. I clean workdir and and don't add "set_prefix", it works now. Thanks

Sorry, Because it's interrupted before, So it restarts and not read params.json. I clean workdir and and don't add "set_prefix", it works now. Thanks